mpi4py-fft 1.0.1

mpi4py-fft

mpi4py-fft is a Python package for computing Fast Fourier Transforms (FFTs). Large arrays are distributed and communications are handled under the hood by MPI for Python (mpi4py). To distribute large arrays we are using a new and completely generic algorithm that allows for any index set of a multidimensional array to be distributed. We can distribute just one index (a slab decomposition), two index sets (pencil decomposition) or even more for higher-dimensional arrays.

In mpi4py-fft there is also included a Python interface to the FFTW library. This interface can be used without MPI, much like pyfftw, and even for real-to-real transforms, like discrete cosine or sine transforms.

Further documentation is found at readthedocs.

Installation

The mpi4py-fft package can be installed using:

pip install mpi4py-fft

or, to get the latest version from bitbucket:

pip install git+https://bitbucket.org/mpi4py/mpi4py-fft@master

It can also be built using conda build from the main source directory:

conda build -c conda-forge conf/
conda create --name mpi4py_fft mpi4py_fft --use-local

which will pull in the required dependencies from the conda-forge channel.

Mpi4py-fft depends on Python packages

• mpi4py

• numpy

• cython

• six

and the C-library

• FFTW

Note that mpi4py requires a working MPI installation, with the compiler wrapper mpicc on your search path. All of the above dependencies are available and will be downlaoded through the conda-forge channel if conda is used for installation.

For IO you need to install either h5py or netCDF4 with support for MPI. These libraries are, unfortunately, not compiled with MPI on conda-forge. The two libraries are available, though, for both OSX and linux (h5py-parallel and netcdf4-parallel) from the spectralDNS channel on anaconda cloud.